JP2013054329A - Method of calculating expected degree of progress for supporting efficient learning, and method of configuring learning support system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method of calculating an expected degree of progress for supporting efficient learning, and a method of configuring a learning support system.SOLUTION: Provided is a method of configuring a learning support system which gives a correlation coefficient between learning steps and preliminarily sets a degree of progress, which is expected after learning in the steps. In this system, a method of selecting a next learning step in accordance with a degree of progress in actual learning and a method of setting again an expected degree of progress for re-learning are provided for learners in each of an advanced class, a middle class, and a beginners' class. The method allows a system to be constructed on the basis of indexes given in advance and is a special method capable of learning guidance for supporting efficient learning. This invention can be utilized in, for example, e-learning systems, learning software, and portable learning devices.

Description

The present invention relates to an improvement degree calculation method for efficiently performing learning, and more particularly to a system that supports efficient learning.

Usually, when constructing an e-learning system, modules are arranged in accordance with the experience of the professor to create a learning course. The evaluation for the e-learning system is generally performed by using a test result or a questionnaire survey after learning using the system. In order to build a system that further improves learning efficiency, traditionally, a small test or the like is performed at the evaluation points at the time of learning progress, and the next learning route is dynamically selected by measuring the degree of improvement from immediately before It was a system to do. In other words, in the methods proposed so far, in order to analyze the learner's data after or during use of the system, first the professor empirically builds a learning system and adapts it to the user's demand. Or a system modification method that can select a learning route according to the learning situation at any time.
Therefore, before building a learning support system such as the first e-learning system, the learning transition trend of learners by advanced, intermediate, and elementary level is estimated, and the expected improvement after one step learning is + α, and the correlation coefficient between steps is We propose an “expected transition” with r. We also propose a learning support system design method using expected transitions.

This is a system construction method in which learning efficiency can be intentionally designed in advance by determining the arrangement of modules and the like of the learning support system based on the prescribed α and r by performing the above-mentioned α pre-setting. .
JP 2007-316285 A (A) JP 2006-293102 A (A) JP 2004-233734 A (A) JP 2003-280510 A (A)

Prior to constructing a learning support system such as an e-learning system, the advanced, intermediate, and beginner learners can set the expected improvement α that is expected after learning each step. That is, it is possible to calculate in advance the degree of learning improvement that cannot be evaluated unless the system is used or the system is being used.

Conventionally, a learning support system has been constructed based on the empirical rules of the professor, but by using the α set as described above, it is possible to construct a learning support system in which the achievement level can be specified in advance.

At each step learning, the learning divergence state is grasped from the difference between the above-defined α and the actual learner's improvement α ′ so that the learning state delay due to the convenience of the learner can be detected at an early stage. Sequential grasp and instruction of learning according to effective learning efficiency.

しており、ｙ^（ｉ）は標準正規分布に従う。）が１ステップ進行後のＸ_ｉ＋１において標準化学習値ｙ^{（ｉ＋１）}に変化する時、ｙ^{（ｉ＋１）}は下記（数１）を用いて事前に導出できる。

Learning step _{X i} and _{X i + 1} correlation _r i is between, there is -1 ≦ _{r i} ≦ 1, when the upon reaching from _{X i} to _{X i + 1} is + alpha _i becomes the learning degree of improvement can be expected, the _{X i} Standardized learning value y ^{(i) of} a student

Y ⁽ⁱ⁾ follows a standard normal distribution. ) Changes to the standardized learning value y ^{(i + 1)} at X _{i + 1} after one step has progressed, y ^{(i + 1)} can be derived in advance using the following (Equation 1).

When the standardized learning value y ^{(i + 1)} after one-step learning is predicted using the standardized learning value y ⁽ⁱ⁾ before learning from ^{(Equation 1)} , the 95% confidence interval of the standard normal random number z is [−1. 96, 1.96], the upper limit value and the lower limit value of y ^{(i + 1)} are ⁽ Expression 2) and (Expression 3), respectively.

なお、Ｘ^{（ｉ＋１）}およびＸ^（ｉ）は学習者に期待する平均的学習値を設定する。In advanced / intermediate / beginner level, in order to obtain a sufficient learning effect at the end of one step, it is desirable to satisfy r _i ≧ 0.4, and in particular, when r _i ≧ 0.8, a stable learning effect is obtained. can get. Also, if a sufficiently large r _i cannot be expected between steps, it is more effective to provide a sufficiently large α _i in the beginner's class. Α _i is derived using (Equation 4).

X ^{(i + 1)} and X ⁽ⁱ⁾ set an average learning value expected from the learner.

In the case of an advanced learner who has a difference β _i with respect to the learning improvement degree α _i ′ of the set learning improvement degree α _i in the actual learning, the following learning step includes The learning route (Equation 6) is appropriately selected. The term “appropriately” refers to a case where β _i is observed that exceeds a certain number of times of one or more in a plurality of step learning (Equation 5).

For the advanced learner, when the expected improvement degree α _{i + 1} is set small, detailed learning or advanced learning is provided. For this reason, as a learner's intention confirmation, the message of selecting detailed learning etc. can be displayed, and also a cheering message can be displayed and learning guidance can be performed.

る。

Is alpha _{i '≦} 0 in the actual learning improvement degree For novice learners, and again in step learning from X _i to X _{i + 1.} However, avoid redoing more than a certain number of times or

The

The beginner learner is provided with a learning instruction by displaying a message regarding redoing or repeating step learning and a support message.

If β _i satisfies (Equation 8), it is considered that the learning is proceeding smoothly.

For the smooth learners, a learning message is displayed by displaying a support message indicating that learning is smooth.

With this method, it is possible to design a learning support system that enables efficient learning by preliminarily defining the improvement rate α between learning steps and the correlation r between steps.

Since there is an indication of the degree of learning improvement and the correlation between steps according to the learner's advanced, intermediate and beginner levels, it is possible to mount a personal learning path selection system that enables stable learning transitions according to each level.

By displaying a support message according to the degree of learning improvement in each step learning, automatic learning guidance according to the learning situation of each person can be realized for each learning step.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 shows an embodiment of the learning step configuration unit and its features will be described. FIG. 1 is a diagram illustrating a configuration method of m learning steps arranged in the learning support system unit. This learning step configuration unit includes m step units (A01 to A05).

In each step part (A01 to A05), a correlation coefficient r between steps and a learning improvement degree α are set. For example, between the i-th step portion (A03) and the i + 1-th step portion (A04), the correlation coefficient r _i is set to the i-th step portion (A03), and after the learning of the i-th step portion (A03) starts. The expected improvement degree α _i of learning expected when reaching the (i + 1) th step unit (A04) is set in the i-th step unit (A03) in advance.

FIG. 2 shows an embodiment of the advanced expectation improvement component and its features will be described. 2 is used selectively in the step learning support unit of FIG. 4 and is configured of a plurality of expected improvement degrees α _{i + 1 (j)} for the i + 1th step unit (A04) located inside the learning step configuration unit of FIG. It is a figure explaining the method. This advanced expectation improvement component comprises an advanced i + 1 step part (B01) and an advanced i + 2 step part (B02).

The expected improvement degrees α _{i + 1 (a)} and α _{i + 1 (b)} set in the advanced i + 1th step section (B01) are determined by (Equation 9). A plurality of α _{i + 1 (b)} may be prepared.

FIG. 3 shows an embodiment of the expected improvement unit for beginners, and its features will be described. 3 is also selectively used in the step learning support unit in FIG. 4 as in FIG. 2, and a plurality of expected improvement degrees α _{i (} for the i-th step unit (A03) located inside the learning step configuration unit in FIG. It is a figure explaining the structure method of _j) . The beginner learning improvement configuration unit includes an i-th step unit for beginners (B03) and an i + 1-th step unit for beginners (B04).

The expected improvement degrees α _{i (a)} and α _{i (b)} set in the i-th step section for beginners (B03) are determined by (Equation 10). A plurality of α _{i (a)} and α _{i (b)} may be prepared.

FIG. 4 shows an embodiment of the step learning support unit, and its features will be described. FIG. 4 is a diagram for explaining a method of configuring step learning arranged in the learning support system unit. The step learning support unit includes an i-th learning unit (C01), a termination processing unit (C02), an actual improvement degree determining unit (C03), an iterative number determining unit (C04), an iterative message unit (C05), and an elementary learning selecting unit. (C06), improvement level resetting unit (C07), improvement level difference determination unit (C08), advanced processing unit (C09), support message unit (C10), i + 1th learning selection unit (C11), i + 1th learning unit ( C12).

The i-th learning unit (C01) is a learning content part in the i-th step unit (A03) in FIG. 1, and α _i has been set in advance.

The termination processing unit (C02) stores the learner's actual learning improvement degree α _i ′ output when the i-th learning unit (C01) is terminated.

The actual improvement determination unit (C03) determines whether α _i ′ is positive or negative and performs branch processing for the beginner learner. First, when α _i ′ ≦ 0, the iteration number determination unit (C04) determines the learning number of the i-th learning unit (C01), and when the number is less than a certain number, the iteration message unit (C05) is output. The i-th learning unit (C01) is executed again to perform iterative learning.

In the actual improvement determination unit (C03), if α _i ′> 0, the process branches to the improvement difference determination unit (C08).

て、向上度再設定部（Ｃ０７）に進む。In the iteration number determination unit (C04), when the number of iterations of the i-th learning unit (C01) is a predetermined number or more, the process branches to the elementary learning selection unit (C06). That is, as a measure for preventing the learning from proceeding to the next step with learning remaining, a measure for reducing the learning load is performed for a learner with a large number of iterations. First, a guidance message indicating that a certain number of iterations have been performed is displayed, and a new load amount is defined by (Equation 7). The i-th and i + 1-th step units for beginners in FIG. 3 (B03, B04)

Then, the process proceeds to the improvement degree resetting unit (C07).

対応する初級用ステップ部（Ｂ０３、Ｂ０４）を選択し、これをｉ番目学習部（Ｃ０１）として進む。

Corresponding beginner step units (B03, B04) are selected, and this is advanced as the i-th learning unit (C01).

The improvement degree difference determination unit (C08) determines whether β _i is positive or negative. If (Equation 5) is satisfied, the process branches to the advanced processing unit (C09). If (Equation 8) is satisfied, the process branches to the support message part (C10).

In the advanced processing unit (C09), when a phenomenon satisfying (Equation 5) is observed exceeding a certain number of times of one or more, the expected improvement degree α _{i + 1} for the next i + 1 learning unit (C12) is expressed by (Equation 9). 2 is selected from the α _{i (b)} set in step _(b) , the corresponding advanced step sections (B01, B02) in FIG. 2 are extracted, and the process proceeds to the (i + 1) -th learning section (C12).

In the support message part (C10), a message indicating that the learning is smoothly progressing is displayed, and the process proceeds to the (i + 1) th learning selection part (C11).

In the (i + 1) th learning selection unit (C11), a correlation coefficient r _{i + 1} between the i + 1th learning unit (C12) and the next i + 2nd learning unit and a normal learning improvement degree α _{i + 1} are set, and the i + 1th learning is performed. Proceed to section (C12).

The i + 1-th learning unit (C12) learns the normal i + 1-th step unit (A05) set according to the value of α _{i + 1} already selected.

FIG. 5 shows an embodiment of the learning support system unit according to the present invention, and its features will be described.
The learning support system unit includes a level determination unit (101), a first step unit (102), a second step unit (103), an m-th step unit (104), and a reaching level output unit (105).

The level determination unit (101) performs academic ability determination before the start of learning, and determines the learning improvement level α1 in the _first step unit (102) which is the initial learning step. α ₁ corresponds to the upper grade, intermediate grade, and beginner grade from the results of academic achievement determination, and is set to normal or small, normal, normal or large, respectively.

At first step portion (102), providing a step learning in accordance with the value of alpha ₁ was set.

In the second step portion (103), in accordance with the first step of the immediately preceding (102) results, the learning degree of improvement alpha ₂ is set. Note that α ₂ is set by the step learning support unit in FIG. Therefore, also in the second step portion (103), providing a step learning in accordance with the set alpha ₂ values.

The m-th step unit (104) also provides step learning according to the value of α _m as in the other step units.

The arrival level output unit (105) determines and displays the arrival result.

Learning step component Advanced expectation improvement component Expectation improvement component for beginners Step learning support department Learning support system

101 level determination unit 102 1st step unit 103 2nd step unit 104 mth step unit 105 reaching level output unit A01 1st step unit A02 2nd step unit A03 ith step unit A04 i + 1st step unit A05 mth step unit B01 Advanced i + 1th step part B02 Advanced i + 2th step part B03 Beginning i-th step part B04 Beginning i + 1st step part C01 i-th learning part C02 End processing part C03 Actual improvement determination part C04 Repetition count determination part C05 Repetitive message Part C06 Elementary learning selection part C07 Improvement degree resetting part C08 Improvement degree difference determination part C09 Advanced processing part C10 Support message part C11 i + 1st learning selection part C12 i + 1st learning part

Claims (4)

A learning course design method that presents the constraints necessary for the configuration of modules and the like when constructing a learning support system that enables learners at each level of advanced, intermediate, and beginners to advance learning efficiently. A method for defining a learning effect (learning improvement degree) α after one-step learning composed of a plurality of modules and a correlation coefficient r between learning steps under the above-described constraints. A dynamic learning path selection method for selecting a next learning path after one-step learning in learning in a learning support system configured by using α and r. A learning instruction method using a difference between a theoretical α value and an α ′ value of an actual learner in a learning support system using a dynamic learning route selection method.

Images